LiDAL: Inter-frame Uncertainty Based Active Learning for 3D LiDAR Semantic Segmentation

11/11/2022
by   Zeyu Hu, et al.
0

We propose LiDAL, a novel active learning method for 3D LiDAR semantic segmentation by exploiting inter-frame uncertainty among LiDAR frames. Our core idea is that a well-trained model should generate robust results irrespective of viewpoints for scene scanning and thus the inconsistencies in model predictions across frames provide a very reliable measure of uncertainty for active sample selection. To implement this uncertainty measure, we introduce new inter-frame divergence and entropy formulations, which serve as the metrics for active selection. Moreover, we demonstrate additional performance gains by predicting and incorporating pseudo-labels, which are also selected using the proposed inter-frame uncertainty measure. Experimental results validate the effectiveness of LiDAL: we achieve 95 learning with less than 5 datasets, outperforming state-of-the-art active learning methods. Code release: https://github.com/hzykent/LiDAL.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset